RESEARCH ARTICLE
Timing of Environmental Enrichment Affects Memory in the House Cricket, Acheta domesticus Heather S. Mallory1¤a*, Aaron F. Howard2¤b, Martha R. Weiss1 1 Department of Biology, Georgetown University, Washington, D. C., United States of America, 2 Department of Biology, Northeastern Illinois University, Chicago, Illinois, United States of America ¤a Current address: Science Department, Ravenscroft School, Raleigh, North Carolina, United States of America ¤b Current address: Department of Biology, Franklin & Marshall College, Lancaster, Pennsylvania, United States of America * [email protected]
Abstract
OPEN ACCESS Citation: Mallory HS, Howard AF, Weiss MR (2016) Timing of Environmental Enrichment Affects Memory in the House Cricket, Acheta domesticus. PLoS ONE 11(4): e0152245. doi:10.1371/journal.pone.0152245 Editor: Nigel E. Raine, University of Guelph, CANADA Received: November 16, 2012 Accepted: March 12, 2016 Published: April 8, 2016 Copyright: © 2016 Mallory et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: Funding was provided by Georgetown University's Graduate School of Arts and Sciences, by Georgetown University’s Department of Biology, and by a seed grant from Georgetown University's Center for the Brain Basis of Cognition. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Learning appears to be ubiquitous among animals, as it plays a key role in many behaviors including foraging and reproduction. Although there is some genetic basis for differences in learning ability and memory retention, environment also plays an important role, as it does for any other trait. For example, adult animals maintained in enriched housing conditions learn faster and remember tasks for longer than animals maintained in impoverished conditions. Such plasticity in adult learning ability has often been linked to plasticity in the brain, and studies aimed at understanding the mechanisms, stimuli, and consequences of adult behavioral and brain plasticity are numerous. However, the role of experiences during postembryonic development in shaping plasticity in adult learning ability and memory retention remain relatively unexplored. Using the house cricket (Acheta domesticus) as a model organism, we developed a protocol to allow the odor preference of a large number of crickets to be tested in a short period of time. We then used this new protocol to examine how enrichment or impoverishment at two developmental stages (either the last nymphal instar or young adult) affected adult memory. Our results show that regardless of nymphal rearing conditions, crickets that experienced an enriched rearing condition as young adults performed better on a memory task than individuals that experienced an impoverished condition. Older adult crickets (more than 1 week post adult molt) did not demonstrate differences in memory of the odor task, regardless of rearing condition as a young adult. Our results suggest that environmentally-induced plasticity in memory may be restricted to the young adult stage.
Competing Interests: The authors have declared that no competing interests exist.
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
1 / 13
Timing of Enrichment Affects Memory in an Insect
Introduction Animals can learn a variety of cues, such as odor, color, shape, and pattern, across different behavioral and environmental contexts [1,2]. Learning and memory ability are not fixed traits, but can be affected by the experiences of the animal. For example, adult rats kept in enriched living conditions consisting of extra space, toys, hidden food, and other rats, learn to run a maze more quickly than do those raised in isolation [3]. Despite a rich literature examining the effects of enrichment on adult learning ability (reviewed by [4]), studies examining the extent to which plasticity of learning ability in the adult is affected by experiences in the juvenile stage are relatively few. Enrichment or deprivation may have a greater effect on adult learning and memory if it occurs while the animal is still undergoing post-embryonic development. Studies in mice and cuttlefish suggest that early experiences may not impact adult learning ability, as impoverishment of the early rearing environment did not dampen the response to later environmental enrichment in the adult [5,6]. However, it is known that sensory and motor experiences, as well as stress and interactions with conspecifics that occur from birth well into adulthood, can have important input on the development of the prefrontal cortex in mammals [7]. It is also possible that duration of enrichment may be even more critical than the life stage at which enrichment occurs, as other studies have shown that the duration of early enrichment determines how long its beneficial effects will persist after animals are switched to an isolated condition [8,9]. These somewhat contradictory results leave open the question of whether rearing condition prior to the adult stage can affect plasticity of learning ability and memory retention in the adult. Additionally, the above studies manipulate the rearing environments to extremes, in an understandable effort to reveal differences; isolated individuals are often maintained in 24 hours of dark, with every effort taken to minimize auditory and olfactory stimulations. Although these studies have provided valuable clues about the influence of sensory information on learning and memory, they do not represent realistic differences in sensory experiences that an individual may experience. Insects are useful model organisms, as they share fundamental behaviors and neurobiological mechanisms with vertebrates, but are less expensive, easier to rear in large numbers, and have smaller, relatively simpler, and more tractable nervous systems [10]. Studies of insect learning and memory have led not only to insights about insect behaviors, but have revealed fundamental principles about the processes underlying these behaviors. For example, studies in the fruit fly Drosophila have added to basic knowledge about the genes, neural pathways, and behaviors involved in learning and memory, and as a consequence Drosophila is currently recognized as a model organism for many disciplines of biology (reviewed by [11]). However, Drosophila are holometabolous insects, which means that they undergo dramatic structural and behavioral changes between the juvenile and adult stages; an egg hatches into a worm-like larva, followed by an outwardly quiescent pupal stage, from which emerges a sexually mature, flying adult. This developmental process differs considerably from the continuous, determinate growth characteristic of vertebrate development. In contrast, hemimetabolous insects undergo incomplete metamorphosis, in which the juvenile resembles the adult in structure and behavior, such that experiences of the juvenile are relevant to the adult stage. Therefore, hemimetabolous insects are better models for examining learning and memory across the lifetime of vertebrates than are holometabolous insects. Crickets are hemimetabolous insects, and in the last decade many behavioral and neurobiological studies have highlighted their strengths as a research organism. Gryllus bimaculatus (the field cricket) can learn olfactory [12,13,14,15] and visual cues [16]. Developmental studies in Acheta domesticus, the common house cricket, reveal that the mushroom bodies— paired structures in the insect brain important for olfactory learning and memory—undergo
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
2 / 13
Timing of Enrichment Affects Memory in an Insect
continuous neurogenesis throughout the life of the animal, including the adult stage, and that adult neurogenesis can be stimulated by sensory input [17, 18,19]. Although the function of adult neurogenesis is unclear, it may play an important role in learning and memory in both invertebrates and vertebrates [20]. We chose Acheta domesticus as our model organism, based on its hemimetabolous life cycle, and the fact that it exhibits adult neurogenesis, which may be important for adult brain plasticity in both vertebrates and invertebrates. We first established a new training protocol that allows for large numbers of crickets to be trained and tested in a single day. We then used this new protocol in two experiments. In order to determine whether timing of enrichment has an effect on memory retention in the adult, we housed crickets in enriched, nutritional control, or impoverished conditions at two developmental stages; either the last nymphal instar (experiment 1) or adults less than 1 week postadult molt (experiment 2). We then maintained the adults in one of the housing conditions above for an additional five days before testing their memory of an olfactory learning task. Using this design, we asked two questions: 1) does previous enrichment or impoverishment affect memory in the adult? and, if so, 2) does the timing of enrichment or impoverishment matter? More specifically, does enrichment or impoverishment that occurs during the nymphal stage (while the animal is still developing) affect adult memory more than enrichment or impoverishment that occurs during the young adult stage (once development is complete)?
Materials and Methods Establishing a Training Protocol Four-week old cricket nymphs (Acheta domesticus) which molted to adulthood in roughly one week, were ordered from PetStuff LLC, maintained in a 20 gallon glass terrarium, and provided with water and Fluker’s orange cube cricket diet. Lights were kept on a 14hr daylight schedule and temperature was maintained at 24°C. Twenty-four hours prior to training, female adults were randomly selected from the terrarium and placed individually in 8 cm x 8 cm x 10 cm transparent plastic boxes with water provided. Food was withheld to motivate foraging behavior in the testing arena. We used only female crickets to control for sex differences observed in pilot studies. To determine whether the crickets preferred either of our test odors, we placed 10 individually marked crickets in a 17.5 cm x 31 cm x 18.5 cm (l x w x h) testing arena. The sides of the arena were wrapped in gray paper to block out interfering visual stimuli, and the arena was lit from above by halogen lamps. Two odor sources were placed in the testing arena, 6 cm from the wall on either end. Odor sources consisted of a 6 cm diameter petri dish containing filter paper soaked in 10μl of essential oil (either rosemary or eucalyptus), and with approximately 100 pin-sized holes in the lid to allow volatiles to disperse. A piece of egg carton in the center of the testing arena provided a resting place for the crickets (Fig 1A). To control for any positional effects, the positions of the odor sources were switched halfway through the 1-hour trial. We video-recorded the trials and later scored them to document number of visits and time spent visiting each odor source. We established an unambiguous criterion to count visits: a visit began when a cricket placed either one of its front legs on the top of the Petri dish, and ended when either one of the front legs of the cricket touched the floor of the testing arena. Commercially available essential oils of two plants, rosemary and eucalyptus, were used as odor sources. We chose these odors because they were plant-based and readily available. As the main goal of this experiment was to establish a new training protocol, rather than investigate the odor preferences of crickets, we felt identity of the odor was less important than our ability to increase an individual’s preference for the odor after pairing it with a reward.
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
3 / 13
Timing of Enrichment Affects Memory in an Insect
Fig 1. Testing and Training. A) Screen shot of testing arena from video recording B) Training dishes. doi:10.1371/journal.pone.0152245.g001
Immediately after odor preference testing was completed, crickets were exposed to a randomly chosen odor, rosemary, in association with Fluker’s orange diet by placing a 2mm plastic lid containing 5μl essential oil of rosemary under another 2mm plastic lid containing the food; the upper lid was perforated with holes to allow the odor to permeate the food without coming into direct contact with it (Fig 1B). Training dishes were kept in the cages with the crickets for 24 hours to give individuals time to associate the presence of food with the odor. Post-training odor preferences were assessed the following day as described above, in the same groups in which they were tested for odor preference prior to training. To test whether crickets were habituating or becoming sensitized to the smell of rosemary, 30 adult female crickets were isolated, tested for odor preference as above, exposed to rosemary odor without food for 24 hours in isolated chambers, and then retested for odor preference in the same groups. Average time spent on each odor source during a trial and number of visits to rosemary and eucalyptus before and after training was compared using paired Student’s T-tests in SPSS.
Effect of Timing of Enrichment on Adult Memory All crickets were kept at 24°C on a 14h daylight cycle. Crickets were ordered as four-week old nymphs from PetStuff LLC and maintained communally in a large colony prior to being placed in one of three conditions: Enriched (E), Impoverished (I), or Nutritional Control (NC). 'E' crickets were placed in groups of four in a 8 cm x 8 cm x 10 cm transparent plastic chamber, reared on Fluker’s orange cube diet (which meets the nutritional needs of the crickets) as well as various fruits and grains, and provided with a piece of egg carton for shelter and sticks for climbing. 'I' crickets were placed singly in a 4cm x 4cm x 7cm translucent plastic chamber and reared on a Fluker's orange cube diet only. 'NC' crickets were raised in conditions identical to 'I', but were provided with the same foods as the 'E' crickets to serve as an intermediate control. We chose an intermediate environment where diet was enriched and social experiences were deprived, as we believed the biggest confound in our impoverished condition was the difference in nutrients. A fourth treatment of socially enriched crickets provided with impoverished diet would have isolated the factor of diet more fully, but would have necessitated increasing the number of treatments from 9 to 16 in order to maintain our crossed experimental design. As we were already near the limits of our capacity to track each cricket individually through the experiment and to generate reasonable sample sizes within a treatment, a compromise was made and the fourth treatment was not included. We believe our NC treatment still serves as an informative intermediate condition, one that is rarely included in studies of environmental effects on learning and memory ability. To determine if rearing conditions had an effect on adult memory, we randomly assigned crickets to one of the above treatments at two different life stages. The full experimental design is shown in Fig 2. For experiment 1, hereafter referred to as the Nymph Experiment, last nymphal instar crickets were maintained in their assigned condition until they molted to the adult stage; time in the assigned condition varied, and ranged from three to seven days (median four days) as we could not predict when nymphs would molt to the adult stage. Newly molted
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
4 / 13
Timing of Enrichment Affects Memory in an Insect
Fig 2. Experimental design. Last instar nymphs (Experiment 1: Nymph Experiment) or young adults (Experiment 2: Adult Experiment) were randomly assigned to one of three conditions: enriched (E), nutritional control (NC), or impoverished (I), (see text for details on conditions). Three to seven days later individuals were randomly assigned to either another or the same condition for five days as an adult. doi:10.1371/journal.pone.0152245.g002
adults were then randomly re-assigned to one of the three conditions and maintained there for five additional days. Thus, a cricket could remain in the same condition or be switched to one of the other two conditions. Crickets assigned to remain in the same treatment were nonetheless transferred to a new cage so that all crickets were handled an equal amount. All cages were kept in the same rearing room throughout the experiment. Using the ‘training protocol’ described above, all crickets were given an olfactory memory task on the fifth day after being switched to the second rearing condition. Odor preference testing was followed immediately by ten minutes of training, and post-training preference testing occurred the next day. Crickets were food-deprived between training and testing. Non-trained and trained groups were kept separated during odor preference training, but treatments were mixed within the group of ten crickets. Crickets were marked as in the experiment above to track individuals, and the observer measuring visits was blind to treatment during scoring. Training time was reduced to ten minutes after pilot tests using adult females demonstrated that a reduction in training time still resulted in a trained preference for rosemary (Fig 3). Twenty-four hours prior to odor preference testing, food, but not water, was withheld to motivate search. Experiment 2, hereafter referred to as the Adult Experiment, was designed to determine whether housing conditions at the young adult stage, just after the adult molt when development was presumably complete, would affect memory in older adults. Comparisons between this experiment (Adult Experiment) and the previous experiment (Nymph Experiment) can reveal if the timing of these experiences is important. Young adult crickets (less than three days since adult molt) were assigned to one of the three conditions for three to seven days (in order to replicate the variation in duration of exposure in the Nymph Experiment), after which they were randomly assigned to one of the three conditions for five additional days. All crickets were then given an olfactory memory task in exactly the same manner as in the Nymph Experiment. To take full advantage of our crossed experimental design, we compared memory across treatments with a single metric. Therefore, instead of comparing number of visits or time spent visiting each odor source independently within each treatment, which would test only whether memory was retained within each treatment, we used the change in time spent on rosemary
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
5 / 13
Timing of Enrichment Affects Memory in an Insect
Fig 3. Appetitive conditioning to rosemary. (A, C) Mean ± SE time spent on each odor before and after appetitive conditioning to rosemary. (B, D) Mean ± SE number of visits on each odor before and after appetitive conditioning to rosemary. (A, B) 24 hours of training, (C, D) 10 minutes of training. See Table 2 for details of statistical analysis. doi:10.1371/journal.pone.0152245.g003
(total time on rosemary after training minus total time spent on rosemary before training) as the response variable. This ‘preference index’ was calculated for each individual cricket, and has been used in several cricket studies as a metric of learning ability and memory [12,13,14]. Data were non-normal with heteroscedasticity, and were therefore square root-transformed across all treatments prior to ANOVA. After transformation, the datasets met the assumptions of normality (KS tests: Nymph experiment D = 0.11, p = 0.12; Adult experiment D = 0.12, p = 0.28) and homoscedasticity (Levene’s tests: Nymph experiment F(8,193) = 1.5, p = 0.14; Adult experiment F(8,77) = 1.05, p = 0.40), and ANOVAs were carried out using first condition and second condition as the main effects and time spent in first condition as a covariate, followed by Tukey’s Post-hocs if overall differences were detected. Cohen’s ƒ2 was calculated to measure the effect size of each factor [21]. Statistics were run in R: A Language and Environment for Statistical Computing (R Core Team).
Results Establishing a Training Protocol Crickets learned to associate the odor of rosemary with a reward when provided with 24 hours of training (Fig 3A and 3B, Table 1). Both the average time spent on rosemary (paired Ttest t50 = -4.86, p = 0.0001) and the number of visits to rosemary (paired T-test t50 = -4.0, p = 0.0002) increased after training. The number of visits and time spent on the non-trained odor, eucalyptus, did not change. Similar results were obtained using 10 minutes of training (Fig 3C and 3D). The time spent on rosemary (paired T-test t28 = -4-2.49, p = 0.019) increased after training, and the number of visits to rosemary increased, although this result was statistically borderline (paired T-test t28 = -2.02, p = 0.05). The number of visits and time spent on the non-trained odor, eucalyptus, did not change. Crickets that were maintained in their cages for 24 hours with the smell of rosemary but no food reward did not show a change in time spent on rosemary (paired T-test t20 = 0.318, p = 0.754), number of visits to rosemary (paired T-test
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
6 / 13
Timing of Enrichment Affects Memory in an Insect
Table 1. Paired-t tests showing change in time spent at and number of visits to the odor sources with 24 hours of training and 10 minutes of training. 10 minutes training df
t
p-value
Time on Rosemary
28
-2.489
0.019
Time on Eucalyptus
28
-0.639
0.528
Number of visits to Rosemary
28
-2.018
0.053
Number of visits to Eucalyptus
28
0
1.00
24 hours training Time on Rosemary
50
-4.857
0.0001
Time on Eucalyptus
50
0.369
0.714
Number of visits to Rosemary
50
-4.001
0.001
Number of visits to Eucalyptus
50
-0.508
0.614
doi:10.1371/journal.pone.0152245.t001
t20 = 1.61, p = 0.123), time spent on eucalyptus (paired T-test t20 = -0.277, p = 0.785) or number of visits to eucalyptus (paired T-test t20 = 0.318, p = 0.754) in preference testing.
Effect of Timing of Enrichment on Adult Memory Nymph experiment. Rearing condition at the last nymphal instar had no effect on memory in the young adults (Table 2, F(2,8) = 1.40, p = 0.869). Regardless of whether they had been isolated or enriched as nymphs, crickets that were enriched as young adults increased their preference for rosemary more than crickets isolated as young adults (Table 2 and Fig 4, F(2,8) = 4.776, p = 0.009, Tukey's post-hoc, E vs I, p = 0.0306). Therefore, isolation at the last nymphal instar did not reduce the ability of a young adult to respond to the enriched condition as an adult. Similarly, enrichment at the last nymphal instar did not carry over to the young adult stage, leading to a young adult from the impoverished condition performing as well on the memory task as a young adult in the enriched condition. Young adults in the nutritional control condition did not significantly differ from enriched young adults, but increased their preference for rosemary more than crickets isolated as young adults (Tukey's post-hoc, NC vs I, p = 0.0315) regardless of their housing condition as last instar nymphs. No interactions between first (nymph) and second (young adult) condition were detected (Table 2, F(2,8) = 1.356, p = 0.251). The amount of time spent in first condition as a last instar nymph, which was run as a covariate, did not have a significant effect (Table 2, F(2,8) = 0.343, p = 0.559). Table 2. Summary of statistical analysis run in SPSS. Significant effects are at the p
Timing of Environmental Enrichment Affects Memory in the House Cricket, Acheta domesticus Heather S. Mallory1¤a*, Aaron F. Howard2¤b, Martha R. Weiss1 1 Department of Biology, Georgetown University, Washington, D. C., United States of America, 2 Department of Biology, Northeastern Illinois University, Chicago, Illinois, United States of America ¤a Current address: Science Department, Ravenscroft School, Raleigh, North Carolina, United States of America ¤b Current address: Department of Biology, Franklin & Marshall College, Lancaster, Pennsylvania, United States of America * [email protected]
Abstract
OPEN ACCESS Citation: Mallory HS, Howard AF, Weiss MR (2016) Timing of Environmental Enrichment Affects Memory in the House Cricket, Acheta domesticus. PLoS ONE 11(4): e0152245. doi:10.1371/journal.pone.0152245 Editor: Nigel E. Raine, University of Guelph, CANADA Received: November 16, 2012 Accepted: March 12, 2016 Published: April 8, 2016 Copyright: © 2016 Mallory et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: Funding was provided by Georgetown University's Graduate School of Arts and Sciences, by Georgetown University’s Department of Biology, and by a seed grant from Georgetown University's Center for the Brain Basis of Cognition. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Learning appears to be ubiquitous among animals, as it plays a key role in many behaviors including foraging and reproduction. Although there is some genetic basis for differences in learning ability and memory retention, environment also plays an important role, as it does for any other trait. For example, adult animals maintained in enriched housing conditions learn faster and remember tasks for longer than animals maintained in impoverished conditions. Such plasticity in adult learning ability has often been linked to plasticity in the brain, and studies aimed at understanding the mechanisms, stimuli, and consequences of adult behavioral and brain plasticity are numerous. However, the role of experiences during postembryonic development in shaping plasticity in adult learning ability and memory retention remain relatively unexplored. Using the house cricket (Acheta domesticus) as a model organism, we developed a protocol to allow the odor preference of a large number of crickets to be tested in a short period of time. We then used this new protocol to examine how enrichment or impoverishment at two developmental stages (either the last nymphal instar or young adult) affected adult memory. Our results show that regardless of nymphal rearing conditions, crickets that experienced an enriched rearing condition as young adults performed better on a memory task than individuals that experienced an impoverished condition. Older adult crickets (more than 1 week post adult molt) did not demonstrate differences in memory of the odor task, regardless of rearing condition as a young adult. Our results suggest that environmentally-induced plasticity in memory may be restricted to the young adult stage.
Competing Interests: The authors have declared that no competing interests exist.
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
1 / 13
Timing of Enrichment Affects Memory in an Insect
Introduction Animals can learn a variety of cues, such as odor, color, shape, and pattern, across different behavioral and environmental contexts [1,2]. Learning and memory ability are not fixed traits, but can be affected by the experiences of the animal. For example, adult rats kept in enriched living conditions consisting of extra space, toys, hidden food, and other rats, learn to run a maze more quickly than do those raised in isolation [3]. Despite a rich literature examining the effects of enrichment on adult learning ability (reviewed by [4]), studies examining the extent to which plasticity of learning ability in the adult is affected by experiences in the juvenile stage are relatively few. Enrichment or deprivation may have a greater effect on adult learning and memory if it occurs while the animal is still undergoing post-embryonic development. Studies in mice and cuttlefish suggest that early experiences may not impact adult learning ability, as impoverishment of the early rearing environment did not dampen the response to later environmental enrichment in the adult [5,6]. However, it is known that sensory and motor experiences, as well as stress and interactions with conspecifics that occur from birth well into adulthood, can have important input on the development of the prefrontal cortex in mammals [7]. It is also possible that duration of enrichment may be even more critical than the life stage at which enrichment occurs, as other studies have shown that the duration of early enrichment determines how long its beneficial effects will persist after animals are switched to an isolated condition [8,9]. These somewhat contradictory results leave open the question of whether rearing condition prior to the adult stage can affect plasticity of learning ability and memory retention in the adult. Additionally, the above studies manipulate the rearing environments to extremes, in an understandable effort to reveal differences; isolated individuals are often maintained in 24 hours of dark, with every effort taken to minimize auditory and olfactory stimulations. Although these studies have provided valuable clues about the influence of sensory information on learning and memory, they do not represent realistic differences in sensory experiences that an individual may experience. Insects are useful model organisms, as they share fundamental behaviors and neurobiological mechanisms with vertebrates, but are less expensive, easier to rear in large numbers, and have smaller, relatively simpler, and more tractable nervous systems [10]. Studies of insect learning and memory have led not only to insights about insect behaviors, but have revealed fundamental principles about the processes underlying these behaviors. For example, studies in the fruit fly Drosophila have added to basic knowledge about the genes, neural pathways, and behaviors involved in learning and memory, and as a consequence Drosophila is currently recognized as a model organism for many disciplines of biology (reviewed by [11]). However, Drosophila are holometabolous insects, which means that they undergo dramatic structural and behavioral changes between the juvenile and adult stages; an egg hatches into a worm-like larva, followed by an outwardly quiescent pupal stage, from which emerges a sexually mature, flying adult. This developmental process differs considerably from the continuous, determinate growth characteristic of vertebrate development. In contrast, hemimetabolous insects undergo incomplete metamorphosis, in which the juvenile resembles the adult in structure and behavior, such that experiences of the juvenile are relevant to the adult stage. Therefore, hemimetabolous insects are better models for examining learning and memory across the lifetime of vertebrates than are holometabolous insects. Crickets are hemimetabolous insects, and in the last decade many behavioral and neurobiological studies have highlighted their strengths as a research organism. Gryllus bimaculatus (the field cricket) can learn olfactory [12,13,14,15] and visual cues [16]. Developmental studies in Acheta domesticus, the common house cricket, reveal that the mushroom bodies— paired structures in the insect brain important for olfactory learning and memory—undergo
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
2 / 13
Timing of Enrichment Affects Memory in an Insect
continuous neurogenesis throughout the life of the animal, including the adult stage, and that adult neurogenesis can be stimulated by sensory input [17, 18,19]. Although the function of adult neurogenesis is unclear, it may play an important role in learning and memory in both invertebrates and vertebrates [20]. We chose Acheta domesticus as our model organism, based on its hemimetabolous life cycle, and the fact that it exhibits adult neurogenesis, which may be important for adult brain plasticity in both vertebrates and invertebrates. We first established a new training protocol that allows for large numbers of crickets to be trained and tested in a single day. We then used this new protocol in two experiments. In order to determine whether timing of enrichment has an effect on memory retention in the adult, we housed crickets in enriched, nutritional control, or impoverished conditions at two developmental stages; either the last nymphal instar (experiment 1) or adults less than 1 week postadult molt (experiment 2). We then maintained the adults in one of the housing conditions above for an additional five days before testing their memory of an olfactory learning task. Using this design, we asked two questions: 1) does previous enrichment or impoverishment affect memory in the adult? and, if so, 2) does the timing of enrichment or impoverishment matter? More specifically, does enrichment or impoverishment that occurs during the nymphal stage (while the animal is still developing) affect adult memory more than enrichment or impoverishment that occurs during the young adult stage (once development is complete)?
Materials and Methods Establishing a Training Protocol Four-week old cricket nymphs (Acheta domesticus) which molted to adulthood in roughly one week, were ordered from PetStuff LLC, maintained in a 20 gallon glass terrarium, and provided with water and Fluker’s orange cube cricket diet. Lights were kept on a 14hr daylight schedule and temperature was maintained at 24°C. Twenty-four hours prior to training, female adults were randomly selected from the terrarium and placed individually in 8 cm x 8 cm x 10 cm transparent plastic boxes with water provided. Food was withheld to motivate foraging behavior in the testing arena. We used only female crickets to control for sex differences observed in pilot studies. To determine whether the crickets preferred either of our test odors, we placed 10 individually marked crickets in a 17.5 cm x 31 cm x 18.5 cm (l x w x h) testing arena. The sides of the arena were wrapped in gray paper to block out interfering visual stimuli, and the arena was lit from above by halogen lamps. Two odor sources were placed in the testing arena, 6 cm from the wall on either end. Odor sources consisted of a 6 cm diameter petri dish containing filter paper soaked in 10μl of essential oil (either rosemary or eucalyptus), and with approximately 100 pin-sized holes in the lid to allow volatiles to disperse. A piece of egg carton in the center of the testing arena provided a resting place for the crickets (Fig 1A). To control for any positional effects, the positions of the odor sources were switched halfway through the 1-hour trial. We video-recorded the trials and later scored them to document number of visits and time spent visiting each odor source. We established an unambiguous criterion to count visits: a visit began when a cricket placed either one of its front legs on the top of the Petri dish, and ended when either one of the front legs of the cricket touched the floor of the testing arena. Commercially available essential oils of two plants, rosemary and eucalyptus, were used as odor sources. We chose these odors because they were plant-based and readily available. As the main goal of this experiment was to establish a new training protocol, rather than investigate the odor preferences of crickets, we felt identity of the odor was less important than our ability to increase an individual’s preference for the odor after pairing it with a reward.
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
3 / 13
Timing of Enrichment Affects Memory in an Insect
Fig 1. Testing and Training. A) Screen shot of testing arena from video recording B) Training dishes. doi:10.1371/journal.pone.0152245.g001
Immediately after odor preference testing was completed, crickets were exposed to a randomly chosen odor, rosemary, in association with Fluker’s orange diet by placing a 2mm plastic lid containing 5μl essential oil of rosemary under another 2mm plastic lid containing the food; the upper lid was perforated with holes to allow the odor to permeate the food without coming into direct contact with it (Fig 1B). Training dishes were kept in the cages with the crickets for 24 hours to give individuals time to associate the presence of food with the odor. Post-training odor preferences were assessed the following day as described above, in the same groups in which they were tested for odor preference prior to training. To test whether crickets were habituating or becoming sensitized to the smell of rosemary, 30 adult female crickets were isolated, tested for odor preference as above, exposed to rosemary odor without food for 24 hours in isolated chambers, and then retested for odor preference in the same groups. Average time spent on each odor source during a trial and number of visits to rosemary and eucalyptus before and after training was compared using paired Student’s T-tests in SPSS.
Effect of Timing of Enrichment on Adult Memory All crickets were kept at 24°C on a 14h daylight cycle. Crickets were ordered as four-week old nymphs from PetStuff LLC and maintained communally in a large colony prior to being placed in one of three conditions: Enriched (E), Impoverished (I), or Nutritional Control (NC). 'E' crickets were placed in groups of four in a 8 cm x 8 cm x 10 cm transparent plastic chamber, reared on Fluker’s orange cube diet (which meets the nutritional needs of the crickets) as well as various fruits and grains, and provided with a piece of egg carton for shelter and sticks for climbing. 'I' crickets were placed singly in a 4cm x 4cm x 7cm translucent plastic chamber and reared on a Fluker's orange cube diet only. 'NC' crickets were raised in conditions identical to 'I', but were provided with the same foods as the 'E' crickets to serve as an intermediate control. We chose an intermediate environment where diet was enriched and social experiences were deprived, as we believed the biggest confound in our impoverished condition was the difference in nutrients. A fourth treatment of socially enriched crickets provided with impoverished diet would have isolated the factor of diet more fully, but would have necessitated increasing the number of treatments from 9 to 16 in order to maintain our crossed experimental design. As we were already near the limits of our capacity to track each cricket individually through the experiment and to generate reasonable sample sizes within a treatment, a compromise was made and the fourth treatment was not included. We believe our NC treatment still serves as an informative intermediate condition, one that is rarely included in studies of environmental effects on learning and memory ability. To determine if rearing conditions had an effect on adult memory, we randomly assigned crickets to one of the above treatments at two different life stages. The full experimental design is shown in Fig 2. For experiment 1, hereafter referred to as the Nymph Experiment, last nymphal instar crickets were maintained in their assigned condition until they molted to the adult stage; time in the assigned condition varied, and ranged from three to seven days (median four days) as we could not predict when nymphs would molt to the adult stage. Newly molted
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
4 / 13
Timing of Enrichment Affects Memory in an Insect
Fig 2. Experimental design. Last instar nymphs (Experiment 1: Nymph Experiment) or young adults (Experiment 2: Adult Experiment) were randomly assigned to one of three conditions: enriched (E), nutritional control (NC), or impoverished (I), (see text for details on conditions). Three to seven days later individuals were randomly assigned to either another or the same condition for five days as an adult. doi:10.1371/journal.pone.0152245.g002
adults were then randomly re-assigned to one of the three conditions and maintained there for five additional days. Thus, a cricket could remain in the same condition or be switched to one of the other two conditions. Crickets assigned to remain in the same treatment were nonetheless transferred to a new cage so that all crickets were handled an equal amount. All cages were kept in the same rearing room throughout the experiment. Using the ‘training protocol’ described above, all crickets were given an olfactory memory task on the fifth day after being switched to the second rearing condition. Odor preference testing was followed immediately by ten minutes of training, and post-training preference testing occurred the next day. Crickets were food-deprived between training and testing. Non-trained and trained groups were kept separated during odor preference training, but treatments were mixed within the group of ten crickets. Crickets were marked as in the experiment above to track individuals, and the observer measuring visits was blind to treatment during scoring. Training time was reduced to ten minutes after pilot tests using adult females demonstrated that a reduction in training time still resulted in a trained preference for rosemary (Fig 3). Twenty-four hours prior to odor preference testing, food, but not water, was withheld to motivate search. Experiment 2, hereafter referred to as the Adult Experiment, was designed to determine whether housing conditions at the young adult stage, just after the adult molt when development was presumably complete, would affect memory in older adults. Comparisons between this experiment (Adult Experiment) and the previous experiment (Nymph Experiment) can reveal if the timing of these experiences is important. Young adult crickets (less than three days since adult molt) were assigned to one of the three conditions for three to seven days (in order to replicate the variation in duration of exposure in the Nymph Experiment), after which they were randomly assigned to one of the three conditions for five additional days. All crickets were then given an olfactory memory task in exactly the same manner as in the Nymph Experiment. To take full advantage of our crossed experimental design, we compared memory across treatments with a single metric. Therefore, instead of comparing number of visits or time spent visiting each odor source independently within each treatment, which would test only whether memory was retained within each treatment, we used the change in time spent on rosemary
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
5 / 13
Timing of Enrichment Affects Memory in an Insect
Fig 3. Appetitive conditioning to rosemary. (A, C) Mean ± SE time spent on each odor before and after appetitive conditioning to rosemary. (B, D) Mean ± SE number of visits on each odor before and after appetitive conditioning to rosemary. (A, B) 24 hours of training, (C, D) 10 minutes of training. See Table 2 for details of statistical analysis. doi:10.1371/journal.pone.0152245.g003
(total time on rosemary after training minus total time spent on rosemary before training) as the response variable. This ‘preference index’ was calculated for each individual cricket, and has been used in several cricket studies as a metric of learning ability and memory [12,13,14]. Data were non-normal with heteroscedasticity, and were therefore square root-transformed across all treatments prior to ANOVA. After transformation, the datasets met the assumptions of normality (KS tests: Nymph experiment D = 0.11, p = 0.12; Adult experiment D = 0.12, p = 0.28) and homoscedasticity (Levene’s tests: Nymph experiment F(8,193) = 1.5, p = 0.14; Adult experiment F(8,77) = 1.05, p = 0.40), and ANOVAs were carried out using first condition and second condition as the main effects and time spent in first condition as a covariate, followed by Tukey’s Post-hocs if overall differences were detected. Cohen’s ƒ2 was calculated to measure the effect size of each factor [21]. Statistics were run in R: A Language and Environment for Statistical Computing (R Core Team).
Results Establishing a Training Protocol Crickets learned to associate the odor of rosemary with a reward when provided with 24 hours of training (Fig 3A and 3B, Table 1). Both the average time spent on rosemary (paired Ttest t50 = -4.86, p = 0.0001) and the number of visits to rosemary (paired T-test t50 = -4.0, p = 0.0002) increased after training. The number of visits and time spent on the non-trained odor, eucalyptus, did not change. Similar results were obtained using 10 minutes of training (Fig 3C and 3D). The time spent on rosemary (paired T-test t28 = -4-2.49, p = 0.019) increased after training, and the number of visits to rosemary increased, although this result was statistically borderline (paired T-test t28 = -2.02, p = 0.05). The number of visits and time spent on the non-trained odor, eucalyptus, did not change. Crickets that were maintained in their cages for 24 hours with the smell of rosemary but no food reward did not show a change in time spent on rosemary (paired T-test t20 = 0.318, p = 0.754), number of visits to rosemary (paired T-test
PLOS ONE | DOI:10.1371/journal.pone.0152245 April 8, 2016
6 / 13
Timing of Enrichment Affects Memory in an Insect
Table 1. Paired-t tests showing change in time spent at and number of visits to the odor sources with 24 hours of training and 10 minutes of training. 10 minutes training df
t
p-value
Time on Rosemary
28
-2.489
0.019
Time on Eucalyptus
28
-0.639
0.528
Number of visits to Rosemary
28
-2.018
0.053
Number of visits to Eucalyptus
28
0
1.00
24 hours training Time on Rosemary
50
-4.857
0.0001
Time on Eucalyptus
50
0.369
0.714
Number of visits to Rosemary
50
-4.001
0.001
Number of visits to Eucalyptus
50
-0.508
0.614
doi:10.1371/journal.pone.0152245.t001
t20 = 1.61, p = 0.123), time spent on eucalyptus (paired T-test t20 = -0.277, p = 0.785) or number of visits to eucalyptus (paired T-test t20 = 0.318, p = 0.754) in preference testing.
Effect of Timing of Enrichment on Adult Memory Nymph experiment. Rearing condition at the last nymphal instar had no effect on memory in the young adults (Table 2, F(2,8) = 1.40, p = 0.869). Regardless of whether they had been isolated or enriched as nymphs, crickets that were enriched as young adults increased their preference for rosemary more than crickets isolated as young adults (Table 2 and Fig 4, F(2,8) = 4.776, p = 0.009, Tukey's post-hoc, E vs I, p = 0.0306). Therefore, isolation at the last nymphal instar did not reduce the ability of a young adult to respond to the enriched condition as an adult. Similarly, enrichment at the last nymphal instar did not carry over to the young adult stage, leading to a young adult from the impoverished condition performing as well on the memory task as a young adult in the enriched condition. Young adults in the nutritional control condition did not significantly differ from enriched young adults, but increased their preference for rosemary more than crickets isolated as young adults (Tukey's post-hoc, NC vs I, p = 0.0315) regardless of their housing condition as last instar nymphs. No interactions between first (nymph) and second (young adult) condition were detected (Table 2, F(2,8) = 1.356, p = 0.251). The amount of time spent in first condition as a last instar nymph, which was run as a covariate, did not have a significant effect (Table 2, F(2,8) = 0.343, p = 0.559). Table 2. Summary of statistical analysis run in SPSS. Significant effects are at the p
